A. Field of the Invention
The invention relates to synthetic chemical compositions and methods of testing such compositions to determine which of such compositions inhibit growth of fungi and other microbes. The compositions of the invention may be used to prevent, limit, or otherwise treat fungal damage to agricultural and horticultural crops, particularly to seeds, seedlings and agricultural commodities, including mature plants such as trees. Certain of the compositions and methods of the invention may be used to treat non-plant fungal and bacterial diseases, such as those of animals including man. The compositions of the invention are especially useful where conditions are conducive to fungal disease development and where control of fungal growth is preferably accomplished with compositions which are not toxic to non-fungal cells.
B. Description of the Related Art
A common problem in agriculture is the reduced yield and crop failure that is caused by plant pathogenic microorganisms. For example, in Texas, diseases of rice, soybeans, and cotton caused by plant pathogenic microorganisms reduced state yields by an estimated 15%, 13%, and 27% in 1992, respectively. In addition to field crop losses, disease losses to horticultural plants and forest trees also occurs. Post-harvest fungal contamination may also lead to reduction in yields and the post-harvest values of plant produce. Current strategies for the control of disease include planting of resistant cultivars and use of known pesticides and fungicides.
Unfortunately, many of the most effective antifungal chemicals show undesirable persistence in the environment. In other instances, such chemicals have low specificity as to the target organism. Where non-target organisms are affected by such low specificity chemicals, it is not uncommon to observe undesired biological activity, including human toxicity (teratogenicity, mutagenicity, carcinogenicity, etc.). Public concerns over the use of foreign chemicals in the environment provide a powerful incentive for developing alternative methods of fungal control.
Common agricultural fungal infections include both vascular and non-vascular diseases. Where the diseases are nonvascular, conditions such as damping off and root rot may occur. Damping off (a symptom of pathogenic attack of seed tissue, e.g., fungal attack) occurs as a result of damage to seeds and seedling roots during germination, either before or after emergence from the soil. Seeds experiencing damping off fail to germinate, become soft, shrink, and finally disintegrate. Post-emergence root rot due to pathogen infection of plant tissue is also responsible for large decreases in the viability of cultivated plants.
Control of damping off and root rot has been attempted by breeding resistant plants with variable success. However, completely resistant cultivars have not been developed and microbial diseases remain a major cause of crop loss. This loss is especially evident in wet growth environments or where crops are repeatedly planted in the same fields. Improved cultural practices, while of some value, similarly fail to provide sufficient relief.
For instance, bananas (family Musaceae) are the most important tropical fruit in the world with more than 62 million tons (if plantains are included) produced annually. For adequate growth of these large perennial herbs, a constant high moisture and tropical temperature are required (16.degree.-35.degree. C.). These same conditions are also highly conducive to the major diseases of banana-panama disease, Sigatoka disease, black leaf streak/black Sigatoka, Moko disease, blackhead, and banana bunchy top virus. Panama disease is caused by Fusarium oxysporum f.sp. cubense (races 1, 2, and 4). Black leaf streak/black Sigatoka is caused by Mycosphaerella fijiensis. Moko disease is caused by Pseudomonas solanacearum.
Biological control of fungal diseases in plants has been the subject of prior investigation. See for example, related U.S. Pat. Nos. 4,942,032 and 4,906,611. These patents disclose the production and use of a naturally-occurring antifungal product ("AFP") produced from Pediococcus species to control post-harvest diseases including mucor rot, gray mold, and blue mold in fruit. U.S. Pat. No. 5,244,680 discloses the use of various species of Cryptococcus to prevent postharvest spoilage of fruit. U.S. Pat. No. 5,049,379 discloses the use of a fungicide isolated from Bacillus cereus or use of the organism itself to control infection of certain legumes by Phytophthora megasperma f. sp. medicaginis and Phytophthora megasperma f. sp. glycinea. The use of microbial agonists to inhibit the growth of pathogenic fungi, however, has severe limitations. The antagonistic effects of the control agent are dependent upon the establishment and growth of that agent in the particular ecological niche in which the pathogen is found. Both abiotic and biotic factors can affect such establishment and subsequent growth of the antagonistic microorganism which in turn affects the production of the anti-fungal agent.
Other research has been directed to the isolation of naturally-occurring antifungal compounds from Pseudomonas species. An isolate from Pseudomonas, L-22-64, and yeast, F-43-31, is able to control blue mold on apples. W. J. Janisiewicz, Phytopathology 78:194-198 (1988). The antifungal agent has been identified as pyrrolnitrin. W. J. Janisiewicz and J. Poitmann, Phytopathology 78:1697-1700 (1988). Other examples of naturally-produced compounds which have antifungal activity include: phenazine, phloroglucinol and pyoluteorin. In many cases, the active component of the natural antifungal agents has not been identified nor completely characterized. Since many of these naturally-produced, antifungal agents are poorly characterized at best, the persistence and toxicity of these compounds in the environment is unknown. Furthermore, the fact that these compounds are produced by microbes in the environment suggests that they may have a limited spectrum of antimicrobial activity.
Peptides are effectors of a variety of physiological processes and can act as antimicrobials inhibiting the growth of fungi and other microbial cells. Several cysteine-rich antifungal peptides have been isolated from radish seeds. These peptides, which range in length from 23 to 30 amino acids, have various antifungal activities against Alternaria brasicola, Botrytis cinerea, Fusarium culmorum, Pyricularia oryzae, Fusarium oxysporum, and Verticillium dahliae. Terras et al., FEBS Letters 316, 233 (1993); Terras et al. J. Biol. Chem. 267, 15301 (1992). Cysteine-rich peptides have also been isolated from the seeds of Mirabilis jalapa. These peptides have antifungal activity against Alternaria brassicola, Ascochyta pisi, Botrytis cinerea, Cercospora beticola, Colletotrichum lindemuthianum, Fusarium culmorum, Fusarium oxysporum, Nectria haematocca, Phoma betea, Pyrenophora tritici-repentis, Pyricularia oryzae, Rhizoctonia solani, Verticillium dahliae and Venturia inaequalis. The peptides also inhibited growth of several Gram-positive bacteria. B. P. A. Cammue, J. of Biol. Chem. 267, 2228 (1992). Other cysteine-rich peptides have been isolated from the seeds of Amaranthus caudatus. W. F. Broekaert et al., Biochemistry 31, 4308 (1992). However, the cysteine content of these peptides if produced synthetically will likely result in polymerization of the monomer peptide into aggregates of two or more monomers. These aggregates have unknown anti-fungal activity. Thus, aggregate formation can result in variable levels of activity and specificity.
Other naturally-occurring antifungal peptides have been characterized to a greater degree. A derivative of a dipeptide ("nitropeptin") has been isolated with growth-inhibitory activity against Pyricularia oryzae. The dipeptide was purified from Streptomyces xanthochromogenus. K. Ohba et al., The Journal of Antibiotics 40, 709-713 (1986). Nitropeptin is thought to be a competitive inhibitor of glutamic acid metabolism in protein synthesis. As such, it is not likely to be an effective antimicrobial in nutrient-rich growing conditions or in organisms with glutamic acid biosynthetic capabilities. A naturally-derived cyclic decapeptide, calophycin, also has been shown to exhibit antimicrobial properties. S.-S. Moon, J. Org. Chem. 57, 1097 (1992). The peptide was found to contain several modified amino acids including a D-aspartic acid residue and an N-methyl asparagine residue and a fatty acid residue (2R, 3R, 4S)-3-amino-2-hydroxy-4-methylpalmitic acid!. However, cyclic peptides are difficult to synthesize in high purity and/or at high recovery rates due to the cyclization chemistry. In addition, Bacillus subtilis antifungal peptides have been demonstrated to control brown rot in peaches. C. G. Guelderner et al., Journal of Agricultural and Food Chemistry 36: 366-370 (1988). Naturally-occurring peptides such as nisin as well as naturally-occurring antibiotic acids such as propionic acid have been used in food preservation. However, the environmental stability and toxicities of these naturally-occurring compounds are generally unknown. Furthermore, where such antifungal agents are derived from natural sources, only one or the other stereoisomer will typically demonstrate the desired antifungal activity.
The use of natural antifungal products isolated in commercial quantity from microorganisms is limited in usefulness due in large part to purification problems. Large scale cell culture of the antifungal agent producing microorganism is required for the purification of the antifungal product. In many instances, the cultural isolate responsible for the production of the antifungal agent is not an isolate which is easily batch-cultured or it is entirely incapable of batch culturing (e.g., obligate pathogens). Furthermore, complicated purification strategies are often required to purify the active product to a reasonable level of homogeneity. A substantial disadvantage to the use of naturally-derived antifungal agents is the potential for co-purification of unwanted microbial byproducts, especially byproducts which are undesirably toxic. In many cases, these factors lead to high production costs and make large scale isolation of antifungal products from natural isolates impractical. Purifications may be even more difficult where racemized mixtures are possible where only a single stereoisomer is active, or where disulfide linkages are possible between peptide monomers.
The search for antimicrobial products for agricultural use has lagged behind the search for such products in human medicine. For instance, synthetic peptides have emerged as useful research tools in the development of vaccines in biomedical research. Pinilla, C., et al. Vaccines 92,25 (1992). Such peptides have been used to resolve details of antigen-antibody interactions (Lerner, R. A. Nature 299,592 (1982)), to map protein products of brain-specific genes (Gramsch, C., et al. Neurochem. 40:1220 (1983)), and to prepare optimal analogs of biologically active peptides (Cull, M. G., et al. Proc. Natl. Acad. Sci. USA 89,865 (1992)), Furka, A., et al. Int. J. Pept. Protein Res. 37:487 (1991)) and in microdilution assays for the development of novel antimicrobial peptides (against S. aureus, P. aeruginosa, C. albicans) (Houghten, et al. 1991; Houghten, et al. 1992a).
U.S. Pat. No. 5,254,535 incorporated herein by reference, discloses the use of peptides and peptide derivatives to potentiate the effects of antibiotics used in standard antimicrobial therapies. The peptides used were 14-50 residues in length, requisitely amphiphilic in nature and are predicted to have ion channel-forming characteristics. It is suggested that these peptides may operate to destabilize cell membranes. The peptides disclosed are composed of tetrameric sequences of defined motifs. The motif includes two adjacent hydrophobic residues and at least one basic residue while the remaining residue may be either basic or neutral. The hydrophobic residues are requisitely located together. While the modes of action of certain peptides have been determined (see, e.g., Fiedler et al. 1982; Isono and Suzuki 1979), mechanisms which explain the mode of action and specificity of such peptides have typically not been determined. Where such studies have been conducted in fungal research, initial studies to determine antifungal mode of action of peptides involved a physical examination of mycelia and cells to determine if the peptides could perturb membrane functions responsible for osmotic balance, as has been observed for other peptides (Zasloff, M. 1987. Proc. Natl. Acad. Sci. USA 84:5449-5453). Other potential modes of action could include disruptions of macromolecular synthesis or metabolism.
U.S. Pat. No. 5,126,257 discloses the isolation and use of a naturally-occurring peptide derived from lysed human polymorphonuclear leukocyte extracts. The active agent kills Gram-positive and Gram-negative bacteria. However, its activity against fungi is not known. U.S. Pat. No. 4,725,576 discloses the use of peptides to fight fungal infections. The peptides contain at least 14% histidine. Histidine hexamers are specifically disclosed that control Candida albicans and Streptococcus mutans infections.
Recently developed methods permit the preparation of synthetic peptide combinational libraries ("SPCLs") that are composed of equimolar mixtures of free peptides that can be used with in vitro methods to determine bioactivity (Furka, A., et al. Int. J. Pept. Protein Res. 37:487 (1991), Houghten, R. A., et al. Nature 354:84 (1991), Houghten, R. A., et al. BioTechniques 13:412 (1992). Libraries can consist of D- or L- amino acid stereoisomers or combinations of L- and D- and/or non-naturally-occurring amino acids. Other methods for synthesizing peptides of defined sequence are also known. Similarly, large scale preparative methods are known. Certain recombinant methods for producing peptides are also known. See, e.g., U.S. Pat. No.4,935,351.
However, many antimicrobial agents are as toxic to the host being treated as they are to the target organism. This is especially true of certain antifungal agents, e.g. amphotericin B. Other antimicrobial agents are very rare and/or expensive. Similarly, where possible it is the standard approach in antibiotic therapy to utilize the lowest efficacious level of a given antimicrobial agent possible. For all of these reasons, it would be particularly useful to discover methods and compositions that allow reduction in the concentration of the antimicrobial agent.
Current compositions for controlling pathogen infections, such as fungal infections, have limited use because of their low specificity, human toxicity and persistence in the environment. Even so, due to the inability to sufficiently control fungal growth through breeding resistance and cultural practices, it is necessary to find antifungal agents that do not have such undesirable characteristics. The use of microbes to control fungus growth in the environment is difficult. The use of natural antifungal products isolated in commercial quantity from microorganisms is also limited in usefulness due in large part to purification problems, especially when having to purify racemized mixtures. Moreover, searching for naturally-produced antifungal agents is a very time-consuming process with a very low-probability of success. Even where such naturally-occurring peptides are located, synthesis of the peptide may be problematic (e.g., disulfide formation, high histidine requirements, etc.). Methods and compositions are needed which will provide a means of easily synthesizing and testing for antimicrobial compositions which do not suffer from the same limitations as naturally-occurring peptides.